Suno 5.5 Adds Voice Cloning and Custom Model Training

Suno has released version 5.5 of its AI music generator, introducing three features designed to personalize music creation: Voices, Custom Models, and My Taste.

Voices: Sing With Your Own Voice

The Voices feature, described by Suno as its most-requested community feature, lets Pro and Premier subscribers record or upload their own vocals to use as the singing voice in AI-generated songs. The company implemented a verification system to prevent misuse: users must record a randomly generated speech phrase that is matched against uploaded audio to confirm ownership of the voice.

Voice profiles remain private by default. Suno plans to add a sharing option in the future, with users maintaining full control over their voice data.

Custom Models and My Taste

Custom Models allows Pro and Premier subscribers to fine-tune the v5.5 model using their own music catalog. Users upload personal tracks, and Suno creates a specialized version that learns individual style and applies it to new compositions. Users can create up to three custom models.

My Taste, available to all users, tracks genre and mood preferences over time and adjusts output recommendations accordingly.

Implications and Industry Context

Suno's emphasis on personalization comes as the company faces ongoing legal challenges from major record labels over copyright. The timing is significant: the company indicated these features "lay the foundation for the next generation of music models we're launching with the music industry later this year," suggesting potential licensing agreements ahead.

Competitor Udio has already been largely absorbed by Universal Music Group, creating pressure on Suno to demonstrate legitimate collaboration pathways with rights holders.

What This Means

Suno is shifting its value proposition from generic AI music generation toward personalized creation tied to individual identity—a defensible position that emphasizes user-generated content and style over dataset-dependent generation. The custom model training and voice cloning features address two major use cases: artists wanting AI-assisted production in their signature style, and creators wanting to produce vocal performances without hiring singers. This positions Suno as a creative tool for existing musicians rather than a wholesale replacement for music creation, which could be strategically important as the company navigates copyright disputes.